India AI Impact Summit 2026

ai-for-resilience-powering-innovation-and-efficiency-in-the-energy-transition

9 min read

TRACK:Resilience, Innovation, and Efficiency
SPEAKERS
  • Mr. Abhishek Ranjan, BSES Rajdhani Power Limited (BRPL)
  • Mr. Ashish Khanna, International Solar Alliance (ISA)
  • Mr. Ashish Kumar Goyal, Uttar Pradesh Power Corporation Limited(UPPCL)
  • Mr. Dwijadas Basak, Tata Power Delhi Distribution Ltd
  • Mr. Ghanshyam Prasad, Central Electricity Authority (CEA)
  • Mr. Naveen Srivastava, Power Grid Corporation of India Limited (PGCIL)
  • Mr. Prashant Dangash, Impresa.ai
  • Mr. Reji Kumar Pillai, India Smart Grid Forum (ISGF)
  • Mr. Siddharth Arora, Energiva Ventures Pvt. Ltd.
  • Ms. Disha Khosla, India Smart Grid Forum (ISGF)
  • Ms. Jaqueline Cochran, National Renewable Energy Laboratory (NREL)
  • Ms. Reena Suri, India Smart Grid Forum (ISGF)
SESSION OVERVIEW

As the energy transition accelerates, power systems are being reshaped by renewable integration, decentralised assets, and rising reliability expectations. This session explores how artificial intelligence can strengthen resilience, drive innovation, and improve efficiency across the energy value chain. Through real-world applications and policy perspectives, the discussion will examine AI-enabled forecasting, automation, and decision-making, alongside governance and workforce considerations, to support scalable, secure, and responsible deployment for a sustainable and future-ready energy ecosystem.

VIDEO RECORDING

AI for Resilience: Powering Innovation and Efficiency in the Energy Transition

Abstract

The session examined how artificial intelligence can reinforce the resilience of India’s power system while accelerating the energy transition. An opening keynote highlighted the historic lag of utilities in technology adoption and introduced a newly published handbook containing 174 global AI/ML/AR/VR use‑cases. Subsequent speakers discussed policy gaps, planning‑stage opportunities, procurement hurdles, and the need for a unified AI‑enabled digital roadmap. The ISGF handbook was presented, summarising key insights, performance gains and implementation gaps. A diverse panel then shared concrete experiences—from distribution‑level smart‑meter analytics to transmission‑sector digital substations and international standards work—culminating in a set of recommendations for utilities, regulators, academia and start‑ups.

Detailed Summary

Shivakumar Kalyanaraman (ANRF) opened the session noting that energy is the foundation of AI—all AI workloads ultimately depend on electricity, most of which now comes from the sun. He traced the evolution of utilities from a “laggard” stance in the 20th century to a more experimental posture post‑Y2K, when automation and IT systems began to mature.

He described the AI‑handbook project:

  • A two‑year effort to catalogue AI/ML/robotics use‑cases across the global power sector.
  • A standard template was sent to utilities in 35 countries; only a minority responded, and most shared limited data.
  • The resulting 300‑page handbook (published Nov 2024) lists 174 vetted use‑cases (≈40 % from India).
  • The handbook also maps the evolving policy & standards landscape (AI Act, OECD trustworthy AI, UK energy regulator guidelines, Singapore’s governance framework, Indian BIS standards).

He mentioned that the handbook was peer‑reviewed, released in November 2024, and will be updated later in 2025 with additional use‑cases.

2. Policy & Regulatory Perspective

Ghanshyam Prasad (CEA) delivered the opening address. Key points:

  • Resilience and reliability remain uneven across rural and urban India; urban grids achieve ~50 % renewable integration, but rural reliability lags far behind.
  • Policy gaps impede AI adoption: lengthy tariff‑determination processes, litigation overload in state and central regulators, and ad‑hoc licensing of transmission projects.
  • AI can streamline regulation – automated, transparent tariff‑setting; digital issuance of transmission licences; AI‑driven reduction of litigation through standardized change‑of‑law clauses.
  • Capacity‑building is essential; the CEA is supporting training programmes to elevate AI literacy among regulators and utilities.

3. AI Opportunities in Planning, Operations & Asset Management

Ashish Kumar Goyal (UPPCL) outlined three strategic pillars where AI can add value:

  1. Planning – resource adequacy, transmission, distribution, and procurement planning require optimisation. He called planning the “low‑hanging fruit,” yet noted that existing tools still need massive improvement, especially for long‑term resource adequacy.
  2. Operations – questions of whether the grid is “optimally operated.” He highlighted the need for AI‑driven real‑time dispatch, micro‑grid management, and demand‑response coordination to handle high renewable penetrations (> 50 %).
  3. Asset Management – current practices suffer from over‑/under‑purchasing of assets and sub‑optimal restoration times. AI‑based predictive maintenance can reduce unplanned outages (15‑40 % reduction reported in the handbook) and extend asset life (20‑30 %).

He stressed that without clear policy support AI‑driven tools cannot reach scale, and that AI could also help standardise litigation and automate tariff processes.

4. DISCOM View – Use‑Case Demand & Procurement Hurdles

Abhishek Ranjan (BRPL) spoke from the distribution side:

  • Customer expectations now mirror e‑commerce (instant, transparent service).
  • Use‑case cataloguing – ISGF’s 174‑case handbook is a vital reference, but many DISCOMs still lack a concrete procurement framework for AI services.
  • Procurement pain points – difficulty in valuing AI solutions, lack of appropriate tender formats (e.g., L1 vs QCBS), and risk aversion for large‑ticket contracts (e.g., a GIS‑mapping tender of ₹200 cr vs a rival bid of ₹800 cr).
  • Digital‑maturity roadmap – proposes a unified DISCOM association roadmap with checklists (technical, commercial, procurement, quality, specification).

He announced a new MBA programme for power‑sector professionals (in partnership with IIM Lucknow), the first cohort starting in 2026, aimed at bridging the skill gap.

5. ANRF Brief – RDI Fund & Grant Programme

Shivakumar Kalyanaraman (ANRF) returned to describe the Anusandhan National Research Foundation:

  • RDI fund – a ₹1 lakh crore “patient‑capital” fund (soft‑loan rates 4‑5 % for 10‑15 yr) launched by the PM in November 2023. ₹4 000 crore already disbursed to intermediate fund managers; more allocations expected by Apr‑May 2026.
  • Grant programme – analogous to US National Science Foundation/DARPA, supporting mission‑mode challenges in AI for power (e.g., renewable forecasting, climate‑impacted grid operation).
  • Collaboration model – ANRF acts as a neutral statutory body, partnering with line ministries, research labs (ICMR, DRDO, DBT, MoES), and international entities (MIT, India AI Mission).
  • Focus on deep‑tech – encourages academia‑industry alliances, especially the pairing of experienced retiring engineers with young AI talent.

He invited the ISGF community to bring forward “root‑node problems” for future RDI funding.

6. ISA Perspective – Global Standards & AI Mission

Ashish Khanna (ISA) positioned the International Solar Alliance within the AI ecosystem:

  • Rapid solar growth – 1 000 GW of solar capacity added globally in two years; 40 % of this is decentralised (rooftop, pumps). India’s rooftop share is only 15 % and faces net‑metering bottlenecks.
  • AI‑mission launch – at the summit, ISA unveiled a global AI‑mission for energy focusing on five pillars:
    1. Interoperable standards – alignment with India Energy Stack, ensuring digital twins can be built only after basic metering & SCADA are in place.
    2. Skills & education – creation of an ISA Academy blending AI & electrical‑engineer curricula, leveraging AI‑driven LMS.
    3. Startup ecosystem – support for 70+ start‑ups; promote digital twins, P2P trading pilots, and cross‑border collaborations.
    4. Financing – patient‑capital facilities to de‑risk pilots and procurement.
    5. Use‑case documentation – systematic capture of AI deployments across the power value chain.

He highlighted that no country yet has a data policy that issues digitally signed certificates to every prosumer—a gap ISA intends to address.

7. Handbook Presentation

Disha Khosla (ISGF) delivered a concise walk‑through of the AI/ML/AR/VR handbook:

SectionContent Highlights
Scope174 use‑cases across generation, transmission, distribution, system ops, trading, RPA, SOC. Data drawn from 35 countries; ~40 % Indian cases.
StructureEach case follows: problem statement → solution description → outcomes (KPIs, performance gains). Cost‑benefit details sparse due to limited utility disclosures.
Key Insights• 15‑40 % reduction in unplanned outages.
• 10‑25 % cut in O&M costs.
• 20‑30 % extension of asset life.
• Noticeable improvement in renewable forecasting & grid reliability.
Early‑Stage Opportunities• ERP‑data asset optimisation,
• AR/VR for training,
• Call‑log analytics for revenue protection,
• Smart‑meter data for demand‑side prediction.
Governance & Standards• EU AI Act 2024, OECD trustworthy AI, UK Energy Regulator ethical AI, Singapore AI governance.
• Indian BIS standards referenced.
Gap Analysis1️⃣ Digital‑infrastructure gaps (metering, SCADA).
2️⃣ Policy & regulatory gaps (standardised licences, tariff automation).
3️⃣ Capacity & institutional gaps (skill shortage).
4️⃣ Financing & procurement challenges.
RoadmapDISCOMs classified into three maturity tiers – Initiators, Integrators, Optimisers – with short‑, medium‑, long‑term milestones.
RecommendationsFor policymakers – harmonise standards, incentivise data sharing.
For regulators – adopt transparent, AI‑driven tariff tools.
For utility leadership – adopt use‑case‑first approach, build data‑lake architecture.
For technology providers – deliver modular, scalable solutions with clear ROI.

She concluded that AI is no longer optional; it is a foundational pillar for grid resilience, renewable integration, and operational efficiency.

8. Panel – Real‑World AI Deployments

8.1 Tata Power Delhi Distribution Ltd (TPDDL) – Dwijadas Basak

  • Digital journey (2005‑2025) – two successive 10‑year roadmaps leading to full smart‑meter penetration and a city‑wide data lake.
  • Pilot‑to‑scale methodology – each AI use‑case starts as a pilot, matures, then rolls out enterprise‑wide. Example: predictive analytics on meter‑reading, disconnection/re‑connection data → default‑risk scoring → targeted field visits, reducing O&M costs.
  • AT&C loss reduction – from 53 % (2002) to < 6 % (2024), driven by AI‑enabled asset monitoring and loss‑detection algorithms.
  • Future AI focus – data historian / lake, AI‑driven asset management, network planning, and power‑procurement optimisation as EV penetration rises. Emphasised the “garbage‑in, garbage‑out” principle.

8.2 BSES Rajdhani Power (BRPL) – Abhishek Ranjan

  • Highlighted customer‑experience expectations (instant service, transparent billing) akin to e‑commerce and telecom.
  • Stressed the need for procurement frameworks that can evaluate AI solutions beyond traditional L1/QCBS models.
  • Announced the MBA programme (IIM Lucknow) to upskill utility managers.

8.3 Power Grid Corp. of India (PGCIL) – Naveen Srivastava

  • Traced PGCIL’s evolution from manual operations (1989) to AI‑enabled digital substations (first 765 kV digital substation – “PowerGate”) and digital line‑tunnels (DLL).
  • Data volume – ~40 TB of real‑time sensor data; AI tools used for renewable‑energy management, VMS/PMU integration, asset performance management, robotic inspection.
  • Reported 99.82 % outage reduction via AI‑driven monitoring and predictive analytics; plans to expand AI‑based generative models for grid stability.
  • Emphasised collaborations with academia (COE) and cybersecurity initiatives.

8.4 National Renewable Energy Laboratory – Jacqueline Cochran

  • Presented the global AI‑mission for energy (launched at the summit).
  • Proposed a paradigm shift: rather than conventional interconnection studies, AI could continuously match grid needs with developer proposals, eliminating lengthy interconnection queues.
  • Stressed the importance of interoperable standards and AI‑enabled learning‑management tools for upskilling engineers worldwide.

8.5 Energiva Ventures – Siddharth Arora

  • Offered a venture‑capital view: AI start‑ups need clear, real‑world pilot opportunities to demonstrate ROI.
  • Highlighted the handshake between innovative start‑ups and utilities that possess data but lack AI expertise.

8.6 Impresa.ai – Prashant Dangash

  • Advised utilities to start small: pick a well‑defined problem, leverage existing data, and avoid overwhelming legacy IT stacks.
  • Demonstrated a smart‑meter analytics rollout that now runs 22 AI models for a client, showing that incremental scaling is feasible.

9. Closing Remarks & Logistics

The moderator (Rina) thanked the panel, reminded attendees of the remaining photograph session, and announced that the exhibition hall would stay open until 8 p.m., while the conference rooms would close by 4:20 p.m.

Key Takeaways

  • Comprehensive AI handbook: 174 globally‑sourced use‑cases (≈40 % Indian) demonstrate 15‑40 % outage reduction, 10‑25 % O&M cost cuts, and 20‑30 % asset‑life extension.
  • Policy & regulatory reform is essential: AI can automate tariff setting, streamline licensing, and reduce litigation, but requires transparent, standardized frameworks.
  • Planning, operations, and asset management are the three AI‑impact pillars; planning offers the quickest ROI, while asset‑performance management yields the greatest reliability gains.
  • DISCOM procurement remains a bottleneck; clear AI‑specific tender structures and valuation methods are needed to avoid cost‑inflation and risk aversion.
  • Skill gap is acute: utilities must couple senior power engineers with AI‑savvy youth, supported by targeted MBA/skill‑development programmes.
  • ANRF’s RDI fund (₹1 lakh crore) and grant programme provide patient capital and mission‑mode funding for deep‑tech AI innovations in the power sector.
  • ISA’s AI‑mission stresses five pillars—interoperable standards, skills, start‑up ecosystem, financing, and use‑case documentation—to accelerate AI‑driven renewable integration.
  • Real‑world implementations (TPDDL, PGCIL, BRPL) confirm that pilots can scale to enterprise‑wide AI systems when backed by disciplined data‑lake architectures and robust governance.
  • International perspectives (NREL) envision AI‑driven “continuous interconnection platforms” that replace static studies with real‑time grid‑developer matchmaking.
  • Practical advice: start with a narrowly scoped, data‑rich problem; use lightweight models; iterate; and only later integrate with legacy systems.

See Also:

Graph View

Backlinks